CN100505577C - Method and apparatus for utilizing channel state information in a wireless communication system - Google Patents

Abstract

Techniques for transmitting data from a transmitter unit to a receiver unit in a multiple-input multiple-output (MIMO) communication system. In one method, at the receiver unit, a number of signals are received via a number of receive antennas, with the received signal from the transmitter unit. The received signals are processed to derive channel state information (CSI) indicative of characteristics of a number of transmission channels used for data transmission. The CSI is transmitted back to the transmitter unit. At the transmitter unit, the CSI from the receiver unit is received and data for transmission to the receiver unit is processed based on the received CSI.

Description

Method and apparatus for utilizing channel state information in a wireless communication system

technical field

The present invention relates generally to data communications, and more particularly to a novel and improved method and apparatus for utilizing (in whole or in part) channel state information to provide improved performance in wireless communication systems.

Background technique

Wireless communication systems are widely employed to provide different types of communication such as voice, data, and so on. These systems may be based on Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Orthogonal Frequency Division Modulation (OFDM), or some other modulation technique. OFDM systems can provide high performance in some channel environments.

In terrestrial communication systems (such as cellular systems, broadcast systems, multi-channel multi-point distribution systems (MMDS), and others), an RF modulated signal from a transmitter unit may arrive at a receiver unit via multiple transmission paths. The characteristics of a transmission path generally vary over time due to factors such as fading and multipath.

To provide diversity against degrading path effects and improve performance, multiple transmit and receive antennas may be used. If the transmission paths between the transmit and receive antennas are linearly independent (that is, transmissions on one path do not form a linear combination of transmissions on the other path), which is generally true to some extent, then the transmitted signal is correctly received The probability of increases with the number of antennas. In general, diversity increases and performance improves as the number of transmit and receive antennas increases.

A Multiple-Input Multiple-Output (MIMO) communication system uses multiple ( _NT ) transmit antennas and multiple _NR receive antennas for data transmission. A MIMO channel may be decomposed into N _C independent channels, where N _C ≤ min{N _T , _NR }. Each of the _NC independent channels is also referred to as a spatial sub-channel of the MIMO channel and corresponds to a dimension. MIMO systems can provide improved performance if the additional dimension created by the multiple transmit and receive antennas is used.

There is therefore a need in the art for a technique to utilize channel state information (CSI) to take advantage of the additional dimensionality created by MIMO systems to provide improved system performance.

Contents of the invention

Aspects of this disclosure provide techniques to process received signals within a multiple-input multiple-output (MIMO) communication system to recover transmitted signals and estimate characteristics of the MIMO channel. Different receiver processing schemes may be used to derive channel state information (CSI) that characterizes the transmission channel used for data transmission. The CSI is then reported back to the transmitter system and used to adjust signal processing (eg, coding, modulation, etc.). In this way, high performance is obtained depending on the determined channel conditions.

Certain embodiments of the present invention provide a method of transmitting data from a transmitter unit to a receiver unit in a MIMO communication system. According to the method, at the receiver unit a number of signals are received via a plurality of receiver antennas, the signal received from each receiving antenna comprising a combination of one or more signals transmitted from the transmitter unit. The received signal is processed (e.g., by a Channel Correlation Matrix Inverse (CCMI) scheme, an Unbiased Minimum Mean Square Error scheme (UMMSE), or some other receiver processing scheme) to derive channel state information, which specifies The plurality of transmission channel characteristics, wherein the reported channel state information includes an estimate of a signal-to-noise-plus-interference ratio for each of the plurality of transmission channels. Channel state information is encoded and transmitted back to the transmitter unit. At the transmitter unit, channel state information from the receiver unit is received, and data transmitted to the receiver unit is processed according to the received CSI.

The reported CSI may include full CSI or partial CSI. Full CSI includes sufficient full bandwidth characteristics (eg, magnitude and phase over the available bandwidth) of the propagation paths between all transmit and receive antenna pairs. Part of the CSI may include, for example, the signal-to-noise-plus-interference (SNR) of the transport channel. At the transmit unit, the data for each transport channel may be encoded according to the SNR estimate of the transport channel, and the encoded data for each transport channel may be modulated according to the modulation scheme selected by the SNR estimate. For full CSI processing, the modulation symbols are pre-processed according to the received CSI before transmission.

Another aspect of the present invention provides a multiple-input multiple-output communication system, including: a receiver unit, the receiver unit includes: multiple front-end processors, used to receive multiple signals through multiple receiving antennas and process the received signals to provide received modulation symbols; at least one receive multiple-input multiple-output processor coupled to the front-end processor and used to receive and process received modulation symbols to derive channel state information, which indicates multiple channels for data transmission A transmit channel characteristic, and a transmit data processor, is operatively coupled to the receive MIMO processor and configured to process channel state information for sending it back to the transmitter unit. The communication system also includes a transmitter unit including at least one demodulator configured to receive and process one or more signals from the receiver unit to recover transmitted channel state information, and a transmit data processor configured to Data for transmission to the receiver unit is processed based on recovered channel state information, wherein the reported channel state information includes an estimate of a signal-to-noise-plus-interference ratio for each of a plurality of transmission channels.

Still another aspect of the present invention provides a receiver unit in a multiple-input multiple-output communication system, comprising: a plurality of front-end processors configured to receive a plurality of transmitted signals through a plurality of receiving antennas, and process the received signals to provide received modulation symbols; a filter operatively coupled to a plurality of front-end processors and configured to filter the received modulation symbols according to a first matrix to provide filtered modulation symbols, wherein the first matrix represents the Estimation of channel characteristics between a plurality of transmit antennas and receive antennas for data transmission; a multiplier coupled to the filter and configured to multiply the filtered modulation symbols with a second matrix to provide an estimate of the transmitted modulation symbols , wherein the second matrix is an inverse matrix derived based on the first matrix; a channel quality estimator, coupled to the multiplier and configured to estimate the characteristics of a plurality of transmission channels for data transmission, and provide a number of transmit channel state information for signal-to-noise-and-interference ratio estimation for a plurality of transmission channels between the antenna and the plurality of receive antennas, and a transmit data processor for receiving and processing the channel state information for transmission from the receiver unit.

The present invention also provides methods, systems and devices for realizing various aspects, embodiments and features of the present invention, which will be described in detail as follows.

Description of drawings

The characteristics, properties and advantages of the present invention will become more apparent through the detailed description proposed below in conjunction with the accompanying drawings, in which the same symbols have the same identification, wherein:

1 is a diagram of a multiple-input multiple-output (MIMO) communication system that can implement various aspects and embodiments of the present invention;

2A and 2B are respective block diagrams of embodiments of MIMO transmitter systems capable of partial CSI processing as well as full CSI processing;

3 is a block diagram of an embodiment of a MIMO transmitter system using Orthogonal Frequency Division Modulation (OFDM);

Figure 4 is a block diagram of parts of a MIMO system that can provide different handling for different transmission types and uses OFDM;

Figures 5 and 6 are two corresponding embodiments of receiver systems having multiple ( _NR ) receiving antennas and capable of processing data transmissions according to Channel Correlation Matrix Inverse (CCMI) techniques and Unbiased Minimum Mean Square Error (UMMSE) block diagram of

Figure 7A shows the average throughput of a MIMO system for three receiver processing techniques and different SNR values; and

Figure 7B shows the cumulative probability distribution functions (CDFs) for the three receiver processing techniques generated from the data histograms.

Detailed ways

1 is a diagram of a multiple-input multiple-output (MIMO) communication system 100 that can implement various aspects and embodiments of the present invention. System 100 includes a first system 110 in communication with a second system 150 . System 100 can be used to use a combination of antenna, frequency, and time diversity (described below) to increase spectral efficiency, improve performance, and increase flexibility. In one aspect, system 150 may be used to characterize the communication link and report channel state information (CSI) back to system 110, and system 110 may be used to adjust processing (e.g., coding and modulation) of data to be transmitted based on the reported CSI. ).

Within system 110, data source 112 provides data (i.e., information bits) to transmit (TX) data processor 114, which encodes the data according to a particular coding scheme and interleaves (i.e., reorders) the encoded data according to a particular interleaving scheme. ), and maps the interleaved bits to modulation symbols of one or more transport channels for transmitting data. Interleaving provides time diversity for the coded bits, allows data to be transmitted according to the average signal-to-noise-plus-interference (SNR) ratio of the transmission channel used for data transmission, reduces fading, and removes Correlation between encoded bits. Interleaving may also provide frequency diversity if the coded bits are transmitted on multiple frequency sub-channels. According to an aspect of the present invention, encoding, interleaving, and symbol mapping (or a combination of the above) are implemented according to the full CSI or partial CSI available to the system 110, as indicated in FIG. 1 .

Coding, interleaving, and symbol mapping at transmission system 110 may be accomplished according to a number of schemes. One particular scheme is described in US Patent No. 09776073, entitled "CODING SCHEME FORA WIRELESS COMMUNICATION SYSTEM," filed February 1, 2001, assigned to the assignee of the present invention, and incorporated herein by reference.

MIMO system 100 uses multiple antennas at the transmit and receive ends of the communication link. These transmit and receive antennas may be used to provide various forms of space diversity, including transmit diversity and receive diversity. Space diversity is characterized by the use of multiple transmit antennas and one or more receive antennas. Transmit diversity is characterized by transmitting data over multiple transmit antennas. Typically, additional processing is performed on the data transmitted from the transmit antennas to achieve the desired diversity. For example, data transmitted from different transmit antennas may be delayed or reordered in time, encoded and interleaved at available transmit antennas, and so on. Receive diversity is characterized by the reception of transmitted signals on multiple receive antennas, and diversity is achieved by simply receiving signals via different signal paths.

System 100 may be used in a number of different communication modes, each using antenna, frequency or time diversity, or a combination of the above. Communication modes may include, for example, a "diversity" communication mode as well as a "MIMO" communication mode. The diversity communication mode uses diversity to improve the reliability of the communication link. In the general application of diversity communication mode, which is also referred to as "pure" diversity communication mode, data is transmitted from all available transmit antennas to the receiver system. Pure diversity communication mode may be used when data rate requirements are lower or when SNR is lower or both. The MIMO communication mode uses antenna diversity (ie, multiple transmit antennas and multiple receive antennas) at both ends of the communication link and is generally used to improve reliability and increase the capacity of the communication link. MIMO communication modes may also use frequency and/or time diversity in combination with antenna diversity.

System 100 may also utilize Orthogonal Frequency Division Modulation (OFDM), which effectively divides the operating frequency band into multiple (L) frequency sub-channels (ie, frequency bins). In each time slot (ie a specific time interval depending on the bandwidth of the frequency sub-channels), a modulation symbol may be transmitted on each of the L frequency sub-channels.

System 100 may be used to transmit data over multiple transmission channels. As mentioned above, a MIMO channel may be decomposed into N _C independent channels, N _C ≤ min{N _T , _NR }. Each of the _NC independent channels is also referred to as a spatial sub-channel of the MIMO channel. For MIMO systems that do not use OFDM, there may be only one frequency subchannel and each spatial subchannel may be called a "transport channel". For MIMO systems using OFDM, each spatial subchannel of each frequency subchannel may be referred to as a transport channel. For OFDM systems not operating in MIMO communication mode, there is only one spatial subchannel and each frequency subchannel may be called a transmission channel.

MIMO systems can provide improved performance if the additional dimension created by the multiple transmit and receive antennas is used. This does not necessarily require the CSI to be known at the transmitter, which may increase system efficiency and improve performance when the transmitter is equipped with CSI, which describes the characteristics of the transmission from the transmit antenna to the receive antenna. CSI may be classified as "full CSI" or "partial CSI".

Full CSI includes sufficient characterization of the entire system frequency band (ie, each frequency subchannel) of the propagation path between each transmit-receive antenna pair within the _NT x _NR MIMO matrix. Full CSI processing refers to (1) channel characteristics are available at both the transmitter and receiver, (2) the transmitter calculates the characteristic modulus of the MIMO channel (described below), determines the modulation symbols transmitted with the characteristic modulus, and modulates The symbols are linearly pre-adjusted (filtered), and the pre-adjusted modulation symbols are transmitted, and (3) the receiver implements a complementary operation of linear transmit processing (such as a spatial matched filter) according to the channel characteristics to calculate each transmission channel (ie N _C spatial matched filter coefficients required for each eigenmodulus). Full CSI processing also includes processing the data for each transport channel (eg selecting an appropriate coding and modulation scheme) according to the channel's eigenvalues (described below) to derive modulation symbols.

Part of the CSI may include, for example, the signal-to-noise-plus-interference ratio (SNR) of the transport channel (i.e., the SNR per spatial subchannel for a MIMO system without OFDM, or per frequency per spatial subchannel for a MIMO system with OFDM SNR of the subchannel). Part of the CSI processing may refer to processing the data of each transport channel (eg, selecting an appropriate coding and modulation scheme) according to the channel's SNR.

1, TX MIMO processor 120 receives and processes modulation symbols from TX data processor 114 to provide symbols suitable for transmission over the MIMO channel. The processing implemented by the TX MIMO processor 120 depends on whether full or partial CSI processing is used, as detailed below.

For full CSI processing, the TX MIMO processor 120 may demultiplex and precondition the modulation symbols. For partial CSI processing, TX MIMO processor 120 may simply demultiplex the modulation symbols. Full and partial CSI MIMO processing will be detailed below. For a MIMO system that uses full CSI processing without OFDM, TX MIMO processor 120 provides a stream of preconditioned modulation symbols for each transmit antenna, one preconditioned modulation symbol per slot. Each preconditioned modulation symbol is a linear (and weighted) combination of the _NC modulation symbols for a given time slot of the _NC spatial subchannels, as will be detailed below. For MIMO systems using full CSI processing and OFDM, TX MIMO processor 120 provides a stream of preconditioned modulation symbol vectors for each transmit antenna, each vector comprising L frequency subchannels for a given time slot Preconditioned modulation symbols. For MIMO systems using partial CSI processing but not OFDM, TX MIMO processor 120 provides a stream of modulation symbols for each transmit antenna, one modulation symbol per slot. For MIMO systems using OFDM and partial CSI processing, TX MIMO processor 120 provides a stream of modulation symbol vectors for each transmit antenna, each vector comprising L modulations for L frequency subchannels of a given time slot symbol. For all of the above cases, each modulation symbol or stream of modulation symbol vectors (either preconditioned or not) is received and modulated by a corresponding modulator (MOD) 122 and transmitted via an associated antenna 124 .

In the embodiment shown in FIG. 1 , receiver system 150 includes a number of receive antennas 152 that receive transmitted signals and provide received signals to corresponding demodulators (DEMODs) 154 . Each demodulator 154 implements processing complementary to that at modulator 122 . The modulated symbols from all demodulators 154 are provided to receive (RX) MIMO processor 156 and processed in the manner described below. The received modulation symbols for the transport channel are then provided to an RX data processor 158 which performs processing complementary to TX data processor 114 . In a particular design, RX data processor 158 provides bit values indicative of received modulation symbols, deinterleaves the bit values, and decodes the deinterleaved values to generate decoded bits, which are then provided to data sink 160 . The demapping, deinterleaving and decoding of the received symbols is complementary to the symbol mapping, interleaving and encoding at the transmitter system 110 . The processing of receiver system 150 will be described in detail below.

The spatial subchannels of a MIMO system (or more generally, the transport channel of a MIMO system with or without OFDM) typically experience different link conditions (eg, different fading and multipath effects) and may achieve different SNRs. Accordingly, the capacity of the transmission channel may vary from channel to channel. This capacity may be quantified by the information bit rate (ie the number of information bits per modulation symbol) transmitted on each transport channel for a particular performance level. Also, link conditions typically change over time. As a result of determining the CSI describing link conditions, the supported bit rates for the transport channel also vary over time. To more fully utilize the capacity of the transmission channel, it is possible (typically at the receiving unit) and provided to the transmitting unit so that the processing can be adjusted (or adapted) accordingly. Aspects of the invention provide techniques for determining and utilizing (full or partial) CSI to provide improved system performance.

MIMO Transmitter System with Partial CSI Processing

FIG. 2A is a block diagram of an embodiment of a MIMO transmitter system 110a, which is an embodiment of the transmitter portion of system 110 in FIG. 1 . Transmitter system 110a (not using OFDM) can adjust its processing based on the partial CSI reported by receiver system 150 . System 110a includes (1) a TX data processor 114a that receives and processes information bits to provide modulation symbols and (2) a TXMIMC processor 120a that demultiplexes modulation symbols for _NT transmit antennas.

TX data processor 114a is one embodiment of TX data processor 114 in FIG. 1, and many other designs are possible for TX data processor 114 and are within the scope of the invention. In the particular embodiment shown in FIG. 2A , TX data processor 114a includes encoder 202 , channel interleaver 204 , puncturer 206 , and symbol mapping element 208 . An encoder 202 receives and encodes the information bits according to a particular encoding scheme to provide encoded bits. A channel interleaver 204 interleaves the coded bits according to a particular interleaving scheme to provide diversity. A truncator 206 truncates zero or more interleaved encoded bits to provide the desired encoded bits. A symbol mapping element 208 maps the non-truncated encoded bits into modulation symbols of one or more transport channels for transmitting data.

Although not shown in FIG. 2A for the sake of brevity, pilot data (eg, data of a known pattern) may be encoded and multiplexed with the processed information bits. The processed pilot data may be transmitted (eg in a time division multiplexed fashion) on all or a subset of the set of transport channels used to transmit the information bits. As is known in the art and will be detailed below, pilot data may be used at the receiver to enable channel estimation.

As shown in FIG. 2A , the coding and modulation may be adjusted based on the partial CSI reported by the receiver system 150 . In one embodiment, adaptive coding is achieved by using a fixed base code (e.g., a rate 1/3 Turbo code) and adjusting truncation to obtain the desired code rate, as determined by the SNR of the transmission channel used to transmit the data supported. Alternatively, different coding schemes may be used depending on the reported partial CSI (as indicated by the dashed arrow pointing to block 202). For example, each of the transmission channels may be encoded with a separate code. With this encoding scheme, it is possible to use a sequential "nullization/equalization and interference cancellation" receiver processing scheme to detect and decode the data stream to derive a more reliable estimate of the transmitted data stream. One such receiver processing scheme is described by P.W. Wolniansky, et al in a paper entitled "V-BLAST: An Architecture for Achieving Very High Data Rates over the Rich-Scattering Wireless Channel", Proc. ISSSE-98, Pisa, Italy . Assigned to the assignee of the present invention and incorporated herein by reference.

For each transport channel, the symbol mapping element 208 may be designed to group the untruncated coded bits into groups to form non-binary symbols, and to map the non-binary symbols into corresponding A point within a signal constellation diagram of a particular modulation scheme (eg, QPSK, M-PSK, M-QAM, or some other scheme). Each mapped point corresponds to a modulation symbol. The number of information bits that may be transmitted per modulation symbol for a particular level of performance (eg, one percent packet error rate) depends on the SNR of the transport channel. Therefore, the coding scheme as well as the modulation scheme for each transport channel may be selected according to the reported partial CSI. Channel interleaving may be adjusted according to the reported partial CSI (indicated by the dashed arrow pointing to block 204).

Table 1 lists different coding rate and modulation scheme combinations possible for a certain number of SNR ranges. The bit rate supported by each transport channel may be obtained using any of a number of possible combinations of coding rates and modulation schemes. For example, one information bit per symbol may be obtained by using (1) a coding rate of 1/2, and QPSK modulation (2) a coding rate of 1/3, and 8-PSK modulation, (3) a coding rate of 1 /4, and 16-QAM modulation or some other combination of code rate and modulation scheme. In Table 1, QPSK, 16-QAM, and 64-QAM are used for the listed SNR ranges. Other modulation schemes, such as 8-PSK, 32-QAM, 128-QAM, etc., may also be used and are within the scope of the present invention.

Table 1

SNR range # of information bits/symbols modulation symbol # of encoded bits/symbols encoding rate 1.5-4.4 1 QPSK 2 1/2 4.4-6.4 1.5 QPSK 2 3/4 6.4-8.35 2 16-QAM 4 1/2 8.35-10.4 2.5 16-QAM 4 5/8 10.4-12.3 3 16-QAM 4 3/4 12.3-14.15 3.5 64-QAM 6 7/12 14.15-15.55 4 64-QAM 6 2/3

15.55-17.35 4.5 64-QAM 6 3/4 >17.35 5 64-QAM 6 5/6

The modulation symbols from TX data processor 114a are provided to TX MIMO processor 120a, which is one embodiment of TX MIMO processor 120 in FIG. 1 . Within TX MIMO processor 120a, a demultiplexer 214 demultiplexes received modulation symbols into multiple ( _NT ) modulation symbol streams, one stream for each antenna from which modulation symbols are being transmitted. Each modulation symbol stream is provided to a corresponding modulator 122 . Each modulator 122 converts the modulation symbols to an analog signal, and further amplifies, filters, quadrature modulates, and upconverts the signal to generate a modulated signal suitable for transmission over the wireless link.

If the number of spatial sub-channels is less than the number of available transmit antennas (ie, N _C < N _T ), various schemes may be used for data transmission. In one approach, _NC modulated symbol streams are generated and transmitted on a subset of the available transmit antennas, namely _NC . The remaining _NT - _NC transmit antennas are not used for data transmission. In another approach, The additional degrees of freedom provided by ( _NT - _NC ) additional transmit antennas are used to improve the reliability of data transmission. To this scheme, each of one or more data streams can be encoded, possibly interleaved and distributed between multiple Transmit on Transmit Antennas. Using multiple transmit antennas for data streams increases diversity and improves reliability against degraded path effects.

MIMO Transmitter System with Full CSI Processing

2B is a block diagram of an embodiment of a MIMO transmitter system 110b (not using OFDM) capable of processing data based on the full CSI reported by the receiver system 150. The information bits are encoded, interleaved, and symbol mapped by a TX data processor 114 to generate modulation symbols. Coding and modulation may be adjusted according to the available full CSI reported by the receiver system, and may be performed as described above for MIMO transmitter system 110a.

Within TX MIMO processor 120b, channel MIMO processor 212 demultiplexes the received modulation symbols into multiple (N _C ) modulation symbol streams, one for each spatial subchannel (i.e., characteristic module is) a stream. For full CSI processing, the channel MIMO processor 212 preconditions the _NC modulation symbols at each slot to generate _NT preconditioned modulation symbols, as follows:

$\left[\begin{array}{c}{x}_1\\ x_2\\ m\\ x_{N_T}\end{array}\right]=\left[\begin{array}{cccc}{e}_{11},& e_{12},& & e_{{1N}_c}\\ e_{\mathrm{twenty\; one}},& e_{\mathrm{twenty\; two}},& & e_{{2N}_c}\\ & & & \\ {\mathrm{<figure-callout\; id="1"\; label="e\; N\; T"\; filenames="C02808558E00341.png,C02808558E00351.png"\; state="\{\{state\}\}">e</figure-callout>}}_{N_T},& {\mathrm{<figure-callout\; id="1"\; label="e\; N\; T"\; filenames="C02808558E00341.png,C02808558E00351.png"\; state="\{\{state\}\}">e</figure-callout>}}_{N_T},& & e_{N_T{N}_c}\end{array}\right]\&Center\; Dot;\left[\begin{array}{c}{b}_1\\ b_2\\ m\\ b_{N_C}\end{array}\right]$ Equation (1)

是相应的空间子信道1、2...的调制码元，其中N_C个调制码元的每个可能使用例如M-PSK、M-QAM或一些其它调制方案生成；where b ₁ , b ₂ ... and

are the corresponding spatial subchannels 1, 2... modulation symbols, where each of the _N modulation symbols may be generated using, for example, M-PSK, M-QAM, or some other modulation scheme;

e _ij are the elements of the matrix E of eigenvectors related to the characteristics of the transmission from the transmitting antenna to the receiving antenna; and

是经预调节的调制码元，这可以表达为：

is the preconditioned modulation symbol, which can be expressed as:

${\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="1"\; label="b"\; filenames="C02808558E00341.png,C02808558E00351.png"\; state="\{\{state\}\}">b</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; e</span>}}_{\mathrm{<span\; class="notranslate">\; 11</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; e</span>}}_{\mathrm{<span\; class="notranslate">\; 12</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; C</span>}}}<span\; class="notranslate">\; \&CenterDot;</span>{\mathrm{<span\; class="notranslate">\; e</span>}}_{{\mathrm{<span\; class="notranslate">\; 1</span>}\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; C</span>}}}<span\; class="notranslate">\; ,</span>$

${\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="1"\; label="b"\; filenames="C02808558E00341.png,C02808558E00351.png"\; state="\{\{state\}\}">b</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; e</span>}}_{\mathrm{<span\; class="notranslate">\; twenty\; one</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; e</span>}}_{\mathrm{<span\; class="notranslate">\; twenty\; two</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; C</span>}}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; e</span>}}_{{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; C</span>}}}<span\; class="notranslate">\; ,</span>$

${\mathrm{<span\; class="notranslate">\; x</span>}}_{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; T</span>}}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="1"\; label="b"\; filenames="C02808558E00341.png,C02808558E00351.png"\; state="\{\{state\}\}">b</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="1"\; label="e\; N\; T"\; filenames="C02808558E00341.png,C02808558E00351.png"\; state="\{\{state\}\}">e</figure-callout></span><figure-callout\; id="1"\; label="e\; N\; T"\; filenames="C02808558E00341.png,C02808558E00351.png"\; state="\{\{state\}\}">\; </figure-callout>}}_{{\mathrm{<span\; class="notranslate">\; N</span>}}_{}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; e</span>}}_{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; T</span>}}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; b</span>}}_{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; C</span>}}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; e</span>}}_{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; T</span>}}{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; C</span>}}}$

The eigenvector matrix E may be calculated by the transmitter or provided to the transmitter by the receiver.

For full CSI processing, each preconditioned modulation symbol _xi represents a linear combination of (weighted) modulation symbols of up to _NC spatial subchannels for a particular transmit antenna. The modulation scheme used for each modulation symbol x _i is based on the effective SNR of the eigenmodulus and is proportional to the eigenvalue λ _i (described below). Each of the _NC modulation symbols used to generate each preconditioned modulation symbol may be associated with a different signal constellation. _NT preconditioned modulation symbols generated by channel MIMO processor 212 are demultiplexed by demultiplexer 214 and provided to _NT modulators 122 for each slot.

Full CSI processing is possible depending on available CSI and selected transmit antennas. Full CSI processing may also be selectively and dynamically enabled and disabled. For example, full CSI processing may be enabled for certain data transfers and disabled for others. Full CSI processing may be enabled under certain conditions, such as when the communication link has sufficient SNR.

MIMO Transmitter System with OFDM

Figure 3 is a block diagram of an embodiment of a MIMO transmitter system 110c that uses OFDM and can adjust its processing based on full or partial CSI. The information bits are encoded, interleaved, truncated, and symbol mapped by a TX data processor 114 to generate modulation symbols. Coding and modulation may be adjusted according to available full or partial CSI reported by the receiver system. For MIMO systems with OFDM, modulation symbols may be transmitted from multiple transmit antennas on multiple frequency sub-channels. When operating in pure MIMO communication mode, the transmissions on each frequency sub-channel and from each transmit antenna represent non-duplicated data.

Within the MIMO processor 120c, a demultiplexer (DEMUX) 310 receives and demultiplexes the modulation symbols into a plurality of sub-channel symbol streams S ₁ to S _L , one sub-channel symbol stream for each transmitted code The frequency subchannel of the element.

For full CSI processing, each sub-channel symbol stream is provided to a corresponding sub-channel MIMO processor 312 . Each subchannel MIMO processor 312 demultiplexes the received subchannel symbol stream into multiple (up to N _C ) symbol substreams, one symbol substream for each spatial subchannel used to transmit modulation symbols flow. For full CST processing within OFDM systems, eigenmodes are derived and applied on a per-frequency subchannel basis. Accordingly, each subchannel MIMO processor 312 can precondition up to _Nc modulation symbols according to equation (1) to generate preconditioned modulation symbols. Each preconditioned modulation symbol for a particular transmit antenna for a particular frequency subchannel represents a linear combination of (weighted) modulation symbols for up to _NC spatial subchannels.

For full CSI processing, the (up to) _NT preconditioned modulation symbols per slot generated by each MIMO processor 312 are demultiplexed by the corresponding demultiplexer 314 and provided to ( up to) _NT symbol combiners 316a through 316t. For example, subchannel MIMO processor 312a assigned to frequency subchannel 1 may provide up to _NT preconditioned modulation symbols for frequency subchannel 1 of antennas 1 through _NT . Likewise, subchannel MIMO processor 3121 assigned to frequency subchannel L may provide up to _NT symbols for frequency subchannel L for antennas 1 to _NT .

For partial CSI processing, each subchannel symbol stream, S, is demultiplexed by a corresponding demultiplexer 314 and provided to (up to) _NT symbol combiners 316a through 316t. The processing of the sub-channel MIMO processor 312 is bypassed for part of the CSI processing.

Each combiner 316 receives modulation symbols for up to L frequency subchannels, combines the symbols for each slot into a modulation symbol vector V, and provides the modulation symbol vector to the next processing stage (i.e. modulator 122).

The modulation symbols are thus received and processed by MIMO processor 120c to provide _NT modulation symbol vectors _V1 through _VT , one modulation symbol vector for each transmit antenna. Each modulation symbol vector V occupies a single time slot, and each element of the modulation symbol vector V is associated with a specific frequency subchannel with a unique subcarrier on which the modulation symbol is transmitted. If not operating in a "pure" MIMO communication mode, some of the modulation symbol vectors may have duplicate or redundant information on specific frequency sub-channels for different transmit antennas.

FIG. 3 also shows an embodiment of the demodulator 122 for OFDM. Modulation symbol vectors _V1 through _VT from MIMO processor 120c are provided to corresponding modulators 122a through 122t. In the embodiment shown in FIG. 3 , each modulator 122 includes an inverse fast Fourier transform (IFFT) 320 , a cyclic prefix generator 322 and an upconverter 324 .

IFFT 320 converts each received modulation symbol vector to its time-domain representation (referred to as an OFDM symbol) using an IFFT. IFFT 320 can be designed to perform IFFT on any number of frequency sub-channels (eg, 8, 16, 32, etc.). In one embodiment, for each modulation symbol vector that is converted into an OFDM symbol, cyclic prefix generator 322 repeats a portion of the time domain representation of the OFDM symbol to form a transmission symbol to a particular transmit antenna. The cyclic prefix ensures that the transmitted symbols retain their orthogonality when multipath delay spreads, thus improving the degradation path effects. Implementations of IFFT 320 and cyclic prefix generators are known in the art and will not be described in detail here.

The time-domain representation of each CP generator 322 (i.e., the transmitted symbols for each antenna) is then processed (e.g., converted to analog, modulated, amplified, and filtered) by an upconverter 324 to generate a modulated signal, It is then transmitted from the corresponding antenna 124 .

OFDM modulation is described in detail in the paper entitled "Multicarrier Modulation for Data Transmission: An Idea Whose Time Has Come" by John A.C Bingham, IEEE Journal of Communications, May 1990, which is incorporated herein by reference.

Many different types of transmissions (eg, voice, signaling, data, pilot, etc.) may be transmitted by a communication system. Each of these transfers may require different handling.

Figure 4 is a block diagram of a portion of a MIMO transmitter system capable of providing different processing of different transmission types and using OFDM. The aggregated input data, including all information bits transmitted by system 110d, is provided to demultiplexer 408. The demultiplexer 408 demultiplexes the input data into multiple (K) channel data streams B ₁ through B _k . Each channel data stream may correspond to eg a signaling channel, a broadcast channel, a voice call or a packet data transmission. Each channel data stream is provided to a corresponding TX data processor 114, which encodes the data with the particular coding scheme selected for that channel data stream, interleaves the encoded data according to the particular interleaving scheme, and maps the interleaved bits into Modulation symbols for one or more transport channels that transmit data streams for that channel.

Encoding may be done on a per-transmission basis (ie on a channel stream as shown in Figure 4). However, encoding may also be implemented on aggregate input data (as shown in Figure 1), on multiple channel data streams, on a portion of a channel data stream, across a set of frequency subchannels, across a set of spatial subchannels , across a set of frequency subchannels and spatial subchannels, across each frequency subchannel, over each modulation symbol, or over some other time, space, and frequency unit.

Modulation symbol streams from each TX data processor 114 may be transmitted on one or more frequency subchannels and through one or more spatial subchannels of each frequency subchannel. TX MIMO processor 120d receives a stream of modulation symbols from TX data processor 114 . Depending on the communication mode used for each modulation symbol stream, TX MIMO processor 12Od may demultiplex the modulation symbol stream into multiple subchannel symbol streams. In the embodiment shown in Fig. 4, the modulation symbol stream S ₁ is transmitted on one frequency sub-channel and the modulation symbol stream S _K is transmitted on L frequency sub-channels. The modulated streams for each frequency subchannel are processed by a corresponding subchannel MIMO processor 412, demultiplexed by a demultiplexer 414, and combined by a combiner 416 (e.g., in a manner similar to that described for FIG. 3) to form each Modulation symbol vector of transmit antennas.

In general, the transmitter system encodes and modulates the data for each transmission channel according to information describing the transmission capacity of that channel. This information is generally in the form of full CSI or partial CSI as described above. The full/partial CSI of the transport channel used for data transmission is typically determined at the receiver system and reported back to the transmitter system, which then uses this information to adjust coding and modulation accordingly. The techniques described herein are applicable to multiple parallel transmission channels supported by MIMO, OFDM, or any other communication scheme (eg, a CDMA scheme) capable of supporting multiple parallel transmission channels.

MIMO processing is further described in U.S. Patent Serial No. 09532492, entitled "HIGHEFFICIENCY, HIGH PERFORMANCE COMMUNICATIONS SYSTEM EMPLOYING MULTI-CARRIERMODULATION," filed March 2000, assigned to the assignee of the present invention, and incorporated by reference here.

MIMO receiving system

Aspects of the invention provide techniques for processing received signals within a MIMO system to recover transmitted data and estimate MIMO channel characteristics. The estimated channel characteristics may then be reported back to the transmitter system and used to adjust signal processing (eg, coding, modulation, etc.). In this way, high performance can be achieved depending on the determined channel conditions. Receiver processing techniques described herein include channel correlation matrix inverse (CCMI) techniques, unbiased minimum mean square error (UMMSE) techniques, and full CSI techniques, all of which are detailed below. It is also possible to use other receiver processing techniques and be within the scope of the present invention.

Figure 1 shows a receiver system 150 with multiple ( _NR ) receive antennas and capable of handling data transmissions. Transmit signals from up to _NT transmit antennas are received by _NR antennas 152a through 152r and routed to respective demodulators (DEMODs) 154 (also referred to as front-end processors). For example, receive antenna 152a may also receive multiple transmit signals from multiple transmit antennas, and receive antenna 152r may similarly receive multiple transmit signals. Each demodulator 154 conditions (eg, filters and amplifies) the received signal, downconverts the conditioned signal to an intermediate frequency or baseband, and digitizes the downconverted signal. Each demodulator 154 may also demodulate the digitized samples with received pilots to generate received modulation symbols, which are provided to RXMIMO processor 156 .

Each demodulator 154 also implements complementary processing to that implemented by modulator 122 shown in FIG. 3 if OFDM is used for data transmission. In this case, each demodulator 154 includes an FFT processor (not shown) that generates a transformed representation of the samples and provides a stream of modulation symbol vectors, each vector comprising L of the L frequency subchannels modulation symbols. The streams of modulation symbol vectors from the FFT processors of all demodulators are then provided to a demultiplexer/combiner (not shown in Figure 5), which first combines the modulation symbol vectors from each FFT processor The stream is "channelized" into multiple (up to L) sub-channel symbol streams. Each (up to) L subchannel symbol streams may then be provided to a corresponding RX MIMO processor 156.

For MIMO systems that do not use OFDM, one RX MIMO processor 156 may be used to implement MIMO processing of modulated signals from _NR receive antennas. For MIMO systems using OFDM, one RX MIMO processor 156 may be used to implement MIMO processing of modulated signals from _NR receive antennas for each of the L frequency subchannels used for data transmission.

In a MIMO system with _NT transmit antennas and _NR receive antennas, the received signal at the output of _NR receive antennas can be expressed as:

r = H x + n Equation (2)

where r is the received symbol vector (ie the _NR × 1 vector output from the MIMO channel, measured at the receiving antenna). H is the NR ×N _T channel coefficient matrix giving the channel responses of N _T transmit antennas and _NR receive antennas at a specific moment, and x is the transmit symbol vector (i.e., _{the NT} ×1 vector input to the MIMO channel ), and n is an _NR x 1 vector representing noise plus interference. The received symbol vector r includes _NR modulation symbols for _NR signals received at a particular time by the _NR receive antennas. Likewise, the transmitted symbol vector x includes _NT modulation symbols for the _{NT signals transmitted by the NT transmit} antennas at a particular time.

MIMO Receiver Using CCMI Technology

For CCMI technology, the receiver system first implements the channel matching filtering operation on the received symbol vector r , and the filtered output can be expressed as:

H ^H r = H ^H H x + H ^H n Equation (3)

where the superscript ^"H" stands for transpose and complex conjugation. The square matrix R may be used to represent the product of the channel coefficient matrix H and its conjugate transpose H ^H (ie R=H ^H H).

The channel coefficient matrix H may be derived, for example, from pilot symbols sent with the data. In order to achieve optimal reception and estimate the SNR of the transmission channels, it is most convenient to insert some known symbols in the transmitted data stream and transmit the known symbols on one or more transmission channels. Such known symbols are also called pilot symbols or pilot signals. Methods for estimating a single transmission channel from pilot signals or data transmissions are described in many papers in the art. One such channel estimation method is described in a paper by F. Ling entitled "Optimal Reception, Performance Bound, and Cutoff-Rate Analysis of Referneces-Assisted Coherent CDMA Communications with Applications", IEEE Transaction On Communication. October 1999. This or some other channel estimation method may be extended to matrix form to derive the matrix H of channel coefficients.

An estimate x ′ of the transmitted symbol vector may be obtained by multiplying the signal vector H ^H r by the inverse (or pseudo-inverse) of R, which can be expressed as:

x″=R ^-1 H ^H r

= x + R ^-1 H ^H n

= x + n ′ Equation (4)

From the above equations, it can be concluded that the transmitted symbol vector x may be recovered by matching filtering the received symbol vector r (ie multiplying by the matrix H ^H ), and then multiplying the filtered result by the inverse square matrix R ⁻¹ .

The SNR of the transport channel may be determined as follows. The autocorrelation matrix φ _nn of the noise vector n is first calculated from the received signal. In general, φ _nn is a Hermitian matrix, ie it is complex conjugate symmetric. If the components of the channel noise are uncorrelated and independent and identically distributed (iid), the autocorrelation matrix _φnn of the noise vector n can be expressed as follows:

${\mathrm{\&phi;}}_n={\mathrm{\&sigma;}}_{\mathrm{no}}^{2}I$ as well as

${\mathrm{\&phi;}}_n={\mathrm{\&sigma;}}_{\mathrm{no}}^{2}I$ Equation (5)

是接收到信号的噪声方差。处理后噪声向量n′的自相关矩阵φ_n′n′(即在匹配滤波和与矩阵R^-1左乘后)可表达为：where I is the identity matrix (i.e. ones along the diagonal and zeros elsewhere) and

is the noise variance of the received signal. The autocorrelation matrix φ _n′n′ of the processed noise vector n ′ (i.e. after matched filtering and left multiplication with matrix R ⁻¹ ) can be expressed as:

${\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; \&prime;</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; E.</span>}<span\; class="notranslate">\; [</span>\underset{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; \&prime;</span>{\underset{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; \&prime;</span>}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; ]</span>$

$={\mathrm{\&sigma;}}_{\mathrm{no}}^2{R}^{-1}$ Equation (6)

等于于

其中，是R^-1的第i个对角元素。对不使用OFDM的MIMO系统，第i个元素代表第i个接收天线。且如果使用OFDM，则下标“i”可分解为下标“jk”，其中，“j”代表第j个频率子信道，而“k”代表对应第k个接收天线的第k个空间子信道。From equation (6), the noise variance of the i-th element of the processed noise n ′

equal to

in, is the ith diagonal element of R ^-1 . For MIMO systems that do not use OFDM, the i-th element represents the i-th receive antenna. And if OFDM is used, the subscript "i" can be decomposed into the subscript "jk", where "j" represents the jth frequency subchannel, and "k" represents the kth spatial subchannel corresponding to the kth receiving antenna channel.

For CCMI technology, the SNR of the i-th element (i.e. the i-th element of x ') of the symbol vector received after processing can be expressed as:

${\mathrm{SNR}}_i=\frac{\stackrel{\&OverBar;}{{\left|x_i^{\&prime;}\right|}^2}}{{\mathrm{\&sigma;}}_{\mathrm{no}}^2}$ Equation (7)

的方差平均等于一(1.0)，接收码元向量的SNR可表示为：If the ith transmitted symbol

The variance of is equal to one (1.0) on average, and the SNR of the received symbol vector can be expressed as:

而归一化。The noise variance can be scaled by the ith element of the received symbol vector

And normalized.

The scaled signals from the _NR receive antennas may be summed together to form a combined signal, which may be expressed as:

                       等式(8)

Equation (8)

The SNR of the combined signal, SNR _total, will have a maximum combined SNR equal to the sum of the SNRs of the signals from _NR receiving antennas. The combined SNR might be expressed as:

            等式(9)

Equation (9)

Figure 5 shows an embodiment of the RX MIMO processor 156a that enables the CCMI processing described above. Within RX MIMO processor 156a, the modulation symbols from the _NR receive antennas are multiplexed by a multiplexer 512 to form a stream of received modulation symbol vectors r . The channel coefficient matrix H may be estimated from pilot signals similar to conventional pilot-assisted single and multi-carrier systems, as known in the art. The matrix R is then calculated according to R=H ^H H shown above. The received modulation symbol vectors r are then filtered by matched filter 514, which left-multiplies each vector r by the conjugate transposed channel coefficient matrix ^H , as shown in equation (3). The filtered vector is also left-multiplied by multiplier 516 with the inverse matrix R ^-1 to obtain an estimate x' of the transmitted modulation symbol matrix x , as shown in equation (4).

For some communication modes, subchannel symbol streams from all antennas used for transmission of the channel data stream may be provided to combiner 518, which includes repetition information in time, space, and frequency. The combined modulation symbols x " are then provided to the RX data processor 158. For some other communication modes, the estimated modulation symbols x ' may be provided directly to the RX data processor 158 (not shown in FIG. 5).

The RX MIMO processor 156a thus generates a plurality of symbol streams independently corresponding to the number of transport channels used at the transmitter system. Each symbol stream includes processed modulation symbols, which correspond to modulation symbols before full/partial-CSI processing at the transmitter system. The (processed) symbol stream is then provided to an RX data processor 158 .

Within the RX data processor 158, the processed symbol stream for each modulation symbol is provided to a corresponding demodulation element implementing a demodulation scheme (e.g., M-PSK, M-QAM) that is identical to that at the transmitter system The modulation schemes of the processed transport channels used are complementary. For MIMO communication modes, the modulated data from all assigned demodulators may be decoded independently or multiplexed into one channel data stream and then decoded again, depending on the coding and modulation used at the transmitter unit method. Each channel data stream may then be provided to a corresponding decoder implementing a complementary decoding scheme to that used at the transmitter unit for the channel data stream. The decoded data from each decoder represents an estimate of the transmitted data for that channel data stream.

The estimated modulation symbols x ' and/or the combined modulation symbols x " are also provided to the CSI processor 520, which determines the full or partial CSI of the transmission channel and provides the full/partial CSI to be reported back to the transmitter System 110. For example, the CSI processor 520 may estimate the noise covariance matrix φ _nn' of the ith transmission channel from the received pilot signal, and calculate the SNR according to equations (7) and (9). It may be possible to use a similar Conventional pilot-assisted single and multi-carrier systems estimate SNR, as known in the art. The SNR of transmission channel comprises the part CSI that reports back transmitter system. Modulation symbol is also provided to estimate channel coefficient matrix H accordingly And derive the channel estimator 522 of the square matrix R and the matrix processor 524. The controller 530 is coupled to the RX MIMO processor 156a and the RX data processor 158 and directs the operation of these units.

MIMO Receiver Using UMMSE

可表示为：For the UMMSE technique, the receiver system implementation multiplies the received symbol vector r by the matrix M to derive an initial MMSE estimate of the transmitted symbol vector x

Can be expressed as:

$\underset{\&OverBar;}{\overset{^}{x}}=m\underset{\&OverBar;}{r}$ Equation (10)

和发射码元向量x间(即

)的误差向量e的均方误差最小化。Choose matrix M such that the initial MMSE estimate

and the transmitted symbol vector x (ie

) to minimize the mean square error of the error vector e .

To determine M, the cost function ε can first be expressed as:

ε＝E{ e ^H e }

＝E{[ r ^H M ^H - x ^H ][M r - x ]}

＝E{ r ^H M ^H M r -2Re[ x ^H M r ]+ x ^H x }

To minimize the cost function ε, the derivative of the cost function related to M can be found, and the result can be set to zero, as follows:

$\frac{<span\; class="notranslate">\; \&PartialD;</span>}{<span\; class="notranslate">\; \&PartialD;</span>\mathrm{<span\; class="notranslate">\; m</span>}}\mathrm{<span\; class="notranslate">\; \&epsiv;</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; HH</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; n</span>}}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}$

From E{ xx ^H }=I, E{ rr ^H }=HH ^H +φ _nn , and E{ rx ^H }=H, the following relations are obtained:

2(HH ^H +φ _nn )M ^H ＝2H

Therefore, the matrix M can be expressed as:

M＝H ^H (HH ^H +φ _nn ) ^-1Equation (11)

可以确定为：According to equations (10) and (11), the initial MMSE estimate of the transmitted symbol vector x

can be determined as:

$\underset{<span\; class="notranslate">\; \&OverBar;</span>}{\overset{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; m</span>}\underset{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; r</span>}}$

$=h^h{({\mathrm{HH}}^h+{\mathrm{\&phi;}}_n)}^{-1}\underset{\&OverBar;}{r}$ Equation (12)

的均值而被确定，在加性噪声上取平均，这可以表示为：In order to determine the SNR of the transmission channel of the UMMSE technique, the signal component can first be based on the given time x

is determined by the mean of , averaged over the additive noise, which can be expressed as:

$\mathrm{<span\; class="notranslate">\; E.</span>}<span\; class="notranslate">\; [</span>\underset{<span\; class="notranslate">\; \&OverBar;</span>}{\overset{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}<span\; class="notranslate">\; |</span>\underset{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; E.</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; m</span>}\underset{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; r</span>}}<span\; class="notranslate">\; |</span>\underset{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; ]</span>$

$=h^h{({\mathrm{HH}}^h+{\mathrm{\&phi;}}_n)}^{-1}\mathrm{E.}\left[\underset{\&OverBar;}{r}\right]$

$=h^h{({\mathrm{HH}}^h+{\mathrm{\&phi;}}_n)}^{-1}h\underset{\&OverBar;}{x}$

$=V\underset{\&OverBar;}{x}$

where the matrix V is defined as

V＝{v _ij }

＝MH

＝H ^H (HH ^H +φ _nn ) ^-1 H

use the equation

${<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; HH</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; n</span>}}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; n</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; n</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\mathrm{<span\; class="notranslate">\; h</span>}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; h</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; n</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; n</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}$

Matrix V can be expressed as:

$\mathrm{<span\; class="notranslate">\; V</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; h</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; n</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\mathrm{<span\; class="notranslate">\; h</span>}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; h</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; n</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}$

的第i个元素可被表示为：Initial MMSE estimate

the ith element of can be expressed as:

${\hat{x}}_i=v_{i1}{x}_1+...+v_i{x}_i+...+v_{i_R}{x}_{N_R}$ Equation (13)

的所有元素都不相关且均值为零，则的第i个元素期望值可表示为：if

All elements of are uncorrelated and have a mean of zero, then The expected value of the i-th element of can be expressed as:

$\mathrm{E.}\left[{\hat{x}}_i\right|x]=v_i{x}_i$ Equation (14)

而获得。因此，x的无偏最小均方误差估计

可通过将有偏估计乘以对角矩阵

而获得，如下：As shown in equation (14), is a biased estimate of _xi . The bias can be removed according to the UMMSE technique for improved receiver performance. An unbiased estimate of x _i can be obtained by dividing _vi by

And get. Therefore, the unbiased minimum mean square error estimate of x

can be biased by estimating Multiply the diagonal matrix

And get, as follows:

$\underset{\&OverBar;}{\overset{~}{x}}={\mathrm{D.}}_v^{-1}\underset{\&OverBar;}{\overset{^}{x}}$ Equation (15)

in

${\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; v</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; diag</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; /</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="1"\; label="v"\; filenames="C02808558E00341.png,C02808558E00351.png"\; state="\{\{state\}\}">v</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 11</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; /</span>{\mathrm{<span\; class="notranslate">\; v</span>}}_{\mathrm{<span\; class="notranslate">\; twenty\; two</span>}}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; /</span>{\mathrm{<span\; class="notranslate">\; v</span>}}_{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; R</span>}}{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; R</span>}}}<span\; class="notranslate">\; )</span>$

以及发射的码元向量

之间的误差可以表达为：To determine noise-plus-interference, an unbiased estimate

and the transmitted symbol vector

error between can be expressed as:

$\underset{<span\; class="notranslate">\; \&OverBar;</span>}{\overset{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; e</span>}}}<span\; class="notranslate">\; =</span>\underset{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; v</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\underset{<span\; class="notranslate">\; \&OverBar;</span>}{\overset{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}$

$=\underset{\&OverBar;}{x}-{\mathrm{D.}}_v^{-1}{h}^h{({\mathrm{HH}}^h+{\mathrm{\&phi;}}_n)}^{-1}\underset{\&OverBar;}{r}$

的自相关矩阵可表示为error vector

The autocorrelation matrix of can be expressed as

$=I-{\mathrm{D.}}_v^{-1}{h}^h{({\mathrm{HH}}^h+{\mathrm{\&phi;}}_n)}^{-1}h(1-\frac{1}{2}{\mathrm{D.}}_v^{-1})-(1-\frac{1}{2}{\mathrm{D.}}_v^{-1})h^h{({\mathrm{HH}}^h+{\mathrm{\&phi;}}_n)}^{-1}{\mathrm{HD}}_v^{-1}$

的第i个元素的方差等于u_u。误差向量

的元素是相关的。然而，可能使用足够的交织使得可以忽略误差向量

的元素间的相关性，只有方差影响系统性能。error vector

The variance of the ith element of is equal to u _u . error vector

elements are related. However, it is possible to use enough interleaving such that the error vector

The correlation among the elements, only the variance affects the system performance.

的自相关矩阵可表示为：If the components of the channel noise are uncorrelated and equally distributed, the correlation matrix of the channel noise can be expressed as Equation (5). In this case, the error vector

The autocorrelation matrix of can be expressed as:

${\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; e</span>}}\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; e</span>}}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; x</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; x</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; x</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ]</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; x</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}$

$=u=\left\{u_{\mathrm{ij}}\right\}$

And if the components of the channel noise are uncorrelated, then

$\mathrm{<span\; class="notranslate">\; u</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; v</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; HH</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; n</span>}}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; v</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; v</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; h</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; HH</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; n</span>}}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; HD</span>}}_{\mathrm{<span\; class="notranslate">\; v</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}$

Equation (17)

The SNR of the demodulator output corresponding to the ith transmitted symbol can be expressed as:

${\mathrm{SNR}}_i=\frac{\mathrm{E.}\left[\stackrel{\&OverBar;}{{\left|x_i\right|}^2}\right]}{u_i}$ Equation (18)

If the variance | _xi | ₂ of the processed received symbols xi is equal to one (1.0) on average, the ^SNR of the received symbol vector may be expressed as:

${\mathrm{<span\; class="notranslate">\; SNR</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}$

如等式(10)示出。估计

进一步由乘法器616左乘对角矩阵

得到发射码元向量x的无偏估计

如等式(15)示出。Figure 6 shows an embodiment of the RX MIMO processor 156b, which enables the UMMSE processing described above. Similar to the CCMI method, the matrices H and φ _nn are first estimated from received pilot signals and/or data transmissions. The weighting coefficient matrix M is then calculated according to equation (11). Within RX MIMO processor 156b, the modulation symbols from the _NR receive antennas are multiplexed by a multiplexer 612 to form a received modulation symbol vector r stream. The received modulation symbol vector r is then left multiplied by matrix M by multiplier 614 to obtain an estimate of the transmitted symbol vector x

as shown in equation (10). estimate

Further multiplied by the multiplier 616 to the left of the diagonal matrix

Get an unbiased estimate of the transmitted symbol vector x

as shown in equation (15).

然后提供给RX数据处理器158。对一些其它的通信模式，估计的调制码元可能直接提供给RX处理器158。Likewise, depending on the particular implementation communication mode, the sub-channel symbol streams from all antennas used for channel data stream transmission may be provided to combiner 618, which combines redundant information in time, space and frequency. Combined Modulation Symbols

It is then provided to the RX data processor 158 . For some other communication modes, the estimated modulation symbols Possibly to the RX processor 158 directly.

和/或组合调制码元

还可以提供给CSI处理器620，它确定传输信道的全或部分CSI，并提供全/部分CSI以被报告回发射机系统110。例如，CSI处理器620可能根据等式(16)到(18)估计第i个传输信道的SNR。传输信道的SNR还包括报告回发射机系统的部分CSI。在等式(11)内计算的最优M应该已经最小化误差向量的范数。D_v是根据等式(16)计算的。unbiased estimated modulation symbols

and/or combined modulation symbols

It may also be provided to a CSI processor 620 that determines full or partial CSI for the transport channel and provides full/partial CSI to be reported back to the transmitter system 110 . For example, the CSI processor 620 may estimate the SNR of the i-th transport channel according to equations (16) to (18). The SNR of the transport channel also includes part of the CSI that is reported back to the transmitter system. The optimal M calculated in equation (11) should have minimized the norm of the error vector. D _v is calculated according to equation (16).

MIMO Receiver Using Full CSI Technology

For the full CSI technique, the signal received at the output of _NR receive antennas may be expressed as equation (2) above, namely

r = H x + n

The eigenvector decomposition of the Hermitian matrix formed by the product of the channel matrix and its conjugate transpose can be expressed as:

H ^H H = EΛ.E ^H

Where E is an eigenvector matrix, Λ is a diagonal matrix of eigenvalues, and both dimensions are N _T ×N _T . The transmitter preconditions a set of _NT modulation symbols b using the eigenvector matrix E, as shown in equation (1) above. The transmitted (preconditioned) modulation symbols from the _NT transmit antennas can be expressed as:

x = Eb

Since H ^H H is Hermitian, the eigenvector matrix is unitary. Thus, if the elements of b are idempotent, the elements of x are also idempotent. A received signal may be represented as:

r = HE b + n Equation (19)

The receiver implements a channel matched filter operation followed by multiplication by the right eigenvector. The result of the channel matched filtering and multiplication operations is a vector z that can be expressed as:

z = E ^H H ^H HE b + E ^H H ^H n = Λ b + n ′ Equation (20)

where the covariance of the new noise term can be expressed as:

$\mathrm{E.}\left(\underset{\&OverBar;}{\overset{^}{\mathrm{no}}}{\underset{\&OverBar;}{\overset{^}{\mathrm{no}}}}^h\right)=\mathrm{E.}\left({\mathrm{E.}}^h{h}^h{\underset{\&OverBar;}{n}}^h\mathrm{HE}\right)={\mathrm{E.}}^h{h}^h\mathrm{HE}=\mathrm{\&Lambda;}$ Equation (21)

That is, the noise term is independent of the variance given by the eigenvalues. The SNR of the ith component of z is the ith diagonal element of Λ

Full CSI processing is further described in the aforementioned US Patent Application Serial No. 09532492.

The receiver embodiment shown in Figure 5 may also be used to implement full CSI techniques. The received modulation symbol vectors r are filtered by a matched filter 514, which left-multiplies each vector r by the conjugate transposed channel coefficient matrix ^H , as shown in equation (20) above. The filtered vector is also left multiplied by the multiplier 516 with the right eigenvector ^EH to form an estimate z of the modulation symbol vector b , as shown in equation (20). For the full CSI technique, the matrix processor 524 is used to provide the right eigenvector E ^H . Subsequent processing (eg, by combiner 518 and RX data processor 158) may proceed as described above.

For the full CSI technique, the transmitter unit can select the coding scheme and the modulation scheme (ie signal constellation) for each of the eigenvectors according to the SNR given by the eigenvalues. If the channel conditions are constant at the time interval during which CSI is measured and reported at the receiver and used to precondition the transmission at the transmitter, the performance of the communication system may be equivalent to a set of independent AWGNs with known SNRs channel performance.

Report full or partial CSI back to transmitter system

Using partial CSI (such as CCMI or UMMSE) or full CSI techniques described herein, it is possible to obtain the SNR for each transmitted channel for the received channel. The determined SNR of the transport channel may be reported back to the transmitter system via a back channel. By feeding back the SNR values of the transmitted modulation symbols of the transport channel (i.e. for each spatial subchannel and possibly for each frequency subchannel when using OFDM), it is possible to implement adaptive processing (e.g. adaptive coding and modulation) to improve MIMO Channel usage. For partial CSI feedback techniques, adaptive processing may be obtained without full CSI. For the full CSI feedback technique, enough information (not necessarily explicit eigenvalues and eigenmoduli) is returned to the transmitter to facilitate the calculation of eigenvalues and eigenmodes for each used frequency subchannel.

For CCMI techniques, the SNR value of the received modulation symbol (e.g., ${\mathrm{SNR}}_i=\stackrel{\&OverBar;}{{\left|{x\&prime;}_i\right|}^2}/{\mathrm{\&sigma;}}_{\mathrm{no}}^2,$ or for symbols received on the i-th transmission channel ${\mathrm{SNR}}_i=1/{\mathrm{\&sigma;}}_{\mathrm{no}}^2{r}_i$ ) is fed back to the transmitter. For UMMSE technology, the SNR value of the received modulation symbol (for example, for the symbol SNR _i =E[| _xi |] ² /u _ii or SNR _i =1/u _ii , u _ii calculated by equations (16) and (17)) is fed back to the transmitter. For the full CSI technique, the SNR value of the received modulation symbol (for example, for the symbol received on the i-th transmission channel ${\mathrm{SNR}}_i={\left|z_i\right|}^2/{\mathrm{\&sigma;}}_{\mathrm{no}}^2$ or ${\mathrm{SNR}}_i={\mathrm{\&lambda;}}_i/{\mathrm{\&sigma;}}_{\mathrm{no}}^2,$ where _λii is the eigenvalue of the square matrix R) can be fed back to the transmitter. For full CSI techniques, the characteristic modulus E may be further determined and fed back to the transmitter. For partial and full CSI techniques, the SNR is used to adjust the data processing at the transmitter system. For the full CSI technique, the eigenmode E is further used to precondition the modulation symbols before transmission.

The CSI reported back to the transmitter may be returned either in full or in differential or a combination of the above. In one embodiment, full or partial CSI is reported periodically and different updates are sent back depending on previously sent CSI. As an example of full CSI, the update may be a correction (from the error signal) to the reported characteristic modulus. The eigenvalues generally do not change as quickly as the eigenmoduli, so can be updated at a slower rate. In another embodiment, CSI is only sent when there is a change (eg, if the change exceeds a certain threshold), which may reduce the effective rate of the feedback channel. As an example of partial CSI, SNRs may only be sent back if they have changed (eg differentially). For OFDM systems (with or without MIMO), correlation in the frequency domain may be used to reduce the amount of CSI to be fed back. As an example of an OFDM system using partial CSI, if the SNR of a particular spatial subchannel corresponding to M frequency subchannels is the same, then the SNR and the first and last frequency subchannel for which this condition is true may be reported. It is also possible to use other compression and feedback channel error recovery techniques to reduce the amount of CSI data fed back and are within the scope of the present invention.

Referring back to FIG. 1 , full or partial CSI (eg, channel SNR) determined by RX MIMO processor 156 is provided to TX data processor 162, which processes the CSI and provides processed data to one or more modulators 154. The modulator 154 further conditions the processed data and transmits the CSI back to the transmitter system through the back channel.

At system 110 , the transmitted feedback signal is received by antenna 124 , demodulated by demodulator 122 , and provided to RX data processor 132 . RX data processor 132 performs complementary processing to that performed by TX data processor 162 and recovers the reported full/partial CSI, which is then provided to TX data processor 114 and TX MIMO processor 120 for tailoring the processing by them .

The transmitter system 110 may adjust (ie adapt) its processing according to the full/partial CSI (ie SNR information) from the receiver system 150 . For example, the encoding of each transmission channel may be adjusted so that the information bit rate matches the transmission capability supported by the channel SNR. In addition, the modulation scheme of the transport channel may be selected according to the channel SNR. Other processes, such as interleaving, may also be adapted and are within the scope of the invention. Adjusting the processing of each transport channel according to the determined SNR of each channel enables MIMO systems to achieve high performance (ie high throughput or bit rate for a certain performance level). Adaptive processing can be applied to single-carrier MIMO systems or multi-carrier based MIMO systems (eg MIMO systems using OFDM).

The adjustment of the encoding and the selection of the modulation scheme at the transmitter system may be performed according to a number of techniques, one of which is described in the aforementioned US Patent Application Serial No. 09776073.

Partial (such as CCMI and UMMSE) as well as full CSI techniques are receiver processing techniques that allow MIMO systems to use the additional dimensionality created by using multiple transmit and receive antennas, which is a major advantage for using MIMO. CCMI and UMMSE techniques may allow the same number of modulation symbols to be transmitted per slot as MIMO systems using full CSI, however, other receiver processing techniques may also be used in conjunction with the full/partial CSI feedback techniques described here and in within the scope of the present invention. Analogously, Figures 5 and 6 represent two embodiments of receiver systems capable of handling MIMO transmissions, determining the characteristics of the transmission channel (ie, SNR), and reporting full or partial CSI back to the transmitter system. Other designs and other receiver processing techniques in accordance with the techniques presented here are contemplated and are within the scope of the invention.

It is also possible to directly use partial CSI techniques (such as CCMI and UMMSE techniques) without adaptive processing at the transmitter, only when the total received signal SNR or the total achievable throughput estimated from the SNR is fed back . In one embodiment, the modulation format is determined from the received SNR estimate or the estimated throughput, and the same modulation format is used for all transport channels. This approach may reduce overall system throughput but can greatly reduce the amount of information sent back on the reverse link.

It is possible to achieve system performance improvement using the full/partial CSI feedback technique of the present invention. System throughput with partial CSI feedback can be calculated and compared to that of full CSI feedback. System throughput can be defined as:

$\mathrm{<span\; class="notranslate">\; C</span>}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; C</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; log</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&gamma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>$

where _γi is the SNR per received modulation symbol for partial CSI techniques or the SNR per transport channel for full CSI techniques. The SNR of different processor technologies can be summarized as follows:

For CCMI technology:

For UMMSE techniques: ${\mathrm{\&gamma;}}_i=\frac{1}{u_i},$

For full CSI technology: ${\mathrm{\&gamma;}}_i=\frac{{\mathrm{\&lambda;}}_i}{{\mathrm{\&sigma;}}_{\mathrm{no}}^2},$

Figures 7A and 7B illustrate the performance of a 4x4 MIMO system using partial CSI and full CSI feedback techniques. The results are obtained by computer simulation. In the simulation, each element of the channel system matrix H is modeled as an independent Gaussian random variable with zero mean and unit variance. For each computation, multiple random matrix realizations are generated and the computed throughputs for the realizations are averaged to generate an average throughput.

Figure 7A shows the throughput of a MIMO system for full CSI, partial CSI CCMI and partial CSI U MSE techniques for different SNR values. It can be seen from Fig. 7A that the throughput of the partial CSI UMMSE technique is about 75% of the full SI output at high SNR values and close to the full CSI throughput at low SNR values. The throughput of the partial CSI CCMI technique is about 75% to 90% of the throughput of the partial CSI UMMSE technique at high SNR values, and is about 30% less than that of UMMSE at low SNR values.

Figure 7B Cumulative probability distribution functions (CDFs) for the three techniques generated from the histograms of the data. Figure 7B shows that at an average SNR of 16dB per transmission channel, the throughput is less than 2bps/Hz for about 5% of the cases for the CCMI technique. On the other hand, at the same SNR, the throughput of the UMMSE technique is above 7.5bps/Hz in all cases. Therefore, the UMMSE technique has a lower outage probability than the CCMI technique.

Elements of the transmitter and receiver systems may use one or more digital signal processors (DSPs), application-specific integrated circuits (ASICs), processors, microprocessors, controllers, microcontrollers, field-programmable gate arrays ( FGPA), programmable logic devices, other electronic units, or any combination of the above. Some of the functions and processes described herein may also be implemented in software executing on a processor.

Aspects of the invention can be implemented with a combination of software and applications. For example, calculation of symbol estimates for CCMI and UMMSE techniques and derivation of channel SNR may be implemented according to program code executing on a processor (controllers 530 and 650, respectively, in FIGS. 5 and 6).

The above description of the preferred embodiment enables any person skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments without the exercise of inventiveness. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (38)

1. A method of transmitting data from a transmitter unit to a receiver unit in a multiple-input multiple-output communication system, characterized in that it comprises: In the receiver unit, receiving a plurality of signals via a plurality of receive antennas, wherein the signal received from each receive antenna comprises a combination of one or more signals transmitted from the transmitter unit, processing the received signal to derive channel state information that characterizes a plurality of transmission channels used for data transmission, wherein the derived channel state information includes an estimate of a signal-to-noise-plus-interference ratio for each of the plurality of transmission channels ,as well as transmitting channel state information back to the transmitter unit; and In the transmitter unit, receiving channel state information from the receiver unit, and Data for transmission to the receiver unit is processed based on the received channel state information. 2. The method of claim 1, wherein the processing at the transmitter unit comprises The data for each transport channel is encoded based on a signal-to-noise-plus-interference ratio estimate for the transport channel. 3. The method as claimed in claim 2, characterized in that the data of each transport channel is independently coded based on the estimate of the signal-to-noise-plus-interference ratio of the transport channel. 4. The method of claim 2, wherein encoding comprises encoding the data of the transport channel with a fixed base code, and The truncation of coded bits is adjusted based on a signal-to-noise-plus-interference ratio estimate of the transmission channel. 5. The method of claim 2, wherein the processing at the transmitter unit further comprises The encoded data for each transport channel is modulated according to a modulation scheme selected from an estimate of the signal-to-noise-plus-interference ratio for the transport channel. 6. The method of claim 1, wherein the derived channel state information comprises characteristics of a plurality of transmission channels. 7. The method of claim 1, wherein the derived channel state information indicates eigenmoduli and eigenvalues of a plurality of transmission channels. 8. A method as claimed in claim 7, characterized in that the processing at the transmitter unit comprises encoding the data of the transport channel according to the eigenvalues. 9. A method as claimed in claim 8, characterized in that the data of each transport channel is encoded independently. 10. The method of claim 8, wherein the processing at the transmitter unit further comprises The encoded data of the transport channel is modulated according to a modulation scheme selected according to the eigenvalues to provide modulation symbols. 11. The method of claim 10, wherein the processing at the transmitter unit further comprises preconditioning the modulation symbols before transmission according to the eigenmodulus. 12. The method of claim 1, wherein the channel state information is completely transmitted from the receiver unit. 13. The method of claim 12, wherein the channel state information is periodically transmitted in full from the receiver unit, and wherein updates to the channel state information are transmitted between complete transmissions. 14. The method of claim 1, wherein channel state information is transmitted upon detection of a change in channel characteristics exceeding a certain threshold. 15. The method of claim 7, wherein the channel state information indicating the eigenmodulus and the eigenvalue are transmitted at different update rates. 16. The method of claim 1, wherein the channel state information is derived at the receiver unit from correlation matrix inverse processing. 17. The method of claim 16, wherein the correlation matrix inverse processing at the receiver unit comprises processing the received signal to derive received modulation symbols; filtering the received modulation symbols according to a first matrix to provide filtered modulation symbols, wherein the first matrix represents an estimate of channel characteristics between a plurality of transmit antennas and a plurality of receive antennas for data transmission; multiplying the filtered modulation symbols with a second matrix to provide an estimate of the transmitted modulation symbols, wherein the second matrix is an inverse matrix derived based on the first matrix; and Estimate characteristics of multiple transmission channels for data transmission. 18. The method of claim 17, further comprising: The modulated symbols are estimated to be demodulated according to a particular demodulation scheme to provide demodulated symbols. 19. The method of claim 18, further comprising: The modulated symbols are decoded according to a particular decoding scheme. 20. The method of claim 17, further comprising: The modulation symbol estimates for the redundant transmissions are combined to provide combined modulation symbol estimates. 21. The method of claim 17, further comprising: deriving a matrix of channel coefficients from the received modulation symbols, and where the first matrix is derived from the channel coefficient matrix. 22. The method of claim 21, wherein the matrix of channel coefficients is derived from received modulation symbols corresponding to pilot data. 23. The method of claim 17, wherein the second matrix is an inverse matrix derived from the first matrix. 24. The method of claim 1, wherein the channel state information is derived at the receiver unit according to an unbiased minimum mean square error process. 25. The method of claim 24, wherein said unbiased minimum mean square error processing comprises: processing the received signal to derive received modulation symbols; multiplying the received modulation symbols with the first matrix to provide an estimate of the transmitted modulation symbols; and estimating characteristics of a plurality of transmission channels for data transmission from the received modulation symbols, and where the first matrix is chosen to minimize the mean squared error between the modulation symbol estimate and the transmitted modulation symbol. 26. The method of claim 25, further comprising multiplying the modulation symbol estimate with a second matrix to provide an unbiased estimate of the transmitted modulation symbol, wherein the second matrix is an inverse matrix derived based on the first matrix, and Wherein the characteristics of the transmission channel are estimated based on the estimation of the unbiased modulation symbols. 27. The method of claim 26, further comprising The first matrix is derived from the unbiased modulation symbol estimates and the mean square error between the unbiased modulation symbol estimates and the transmitted modulation symbols is minimized. 28. The method of claim 1, wherein the MIMO system implements quadrature frequency division modulation. 29. The method of claim 28, wherein the processing at the receiver unit or the transmitter unit is performed for each of a plurality of frequency subchannels. 30. A multiple-input multiple-output communication system, characterized in that it comprises: receiver unit, including a plurality of front-end processors for receiving a plurality of signals via a plurality of receive antennas and processing the received signals to provide received modulation symbols, at least one receive multiple-input multiple-output processor coupled to the front-end processor and operable to receive and process received modulation symbols to derive channel state information indicative of characteristics of a plurality of transmission channels used for data transmission, and a transmit data processor operatively coupled to the receive MIMO processor and configured to process channel state information for sending it back to the transmitter unit; and transmitter unit, including at least one demodulator configured to receive and process one or more signals from the receiver unit to recover transmitted channel state information, and a transmit data processor configured to process data for transmission to the receiver unit based on the recovered channel state information, Wherein the derived channel state information includes an estimate of a signal-to-noise-plus-interference ratio for each of the plurality of transmission channels. 31. A receiver unit in a multiple-input multiple-output communication system, characterized in that it comprises a plurality of front-end processors configured to receive a plurality of transmitted signals via a plurality of receive antennas and process the received signals to provide received modulation symbols; a filter operatively coupled to the plurality of front-end processors and configured to filter the received modulation symbols according to a first matrix to provide filtered modulation symbols, wherein the first matrix represents a plurality of transmit antennas for data transmission and Estimation of channel characteristics between receiving antennas; a multiplier coupled to the filter and configured to multiply the filtered modulation symbols with a second matrix to provide an estimate of the transmitted modulation symbols, wherein the second matrix is an inverse derived based on the first matrix matrix; a channel quality estimator coupled to the multiplier and configured to estimate characteristics of the plurality of transmission channels used for data transmission and provide a signal-to-noise-plus-interference ratio indicative of the plurality of transmission channels between the plurality of transmit antennas and the plurality of receive antennas estimated channel state information, and A transmit data processor for receiving and processing channel state information for transmission from the receiver unit. 32. The receiver unit of claim 31, further comprising: A second estimator configured to derive a channel coefficient matrix according to the modulation symbol estimation, and wherein the first matrix is derived according to the channel coefficient matrix. 33. The receiver unit of claim 31 further comprising One or more demodulation elements, each demodulation element configured to receive and demodulate a corresponding stream of modulation symbol estimates according to a particular modulation scheme to provide a stream of demodulated symbols. 34. The receiver unit of claim 33, further comprising One or more decoders, each configured to receive and decode the demodulated symbol stream according to a particular decoding scheme to provide decoded data. 35. The receiver unit of claim 31, wherein the channel quality estimator is further operable to generate channel state information characterizing a plurality of transmission channels. 36. The receiver unit of claim 31, wherein the channel quality estimator is further configured to generate channel state information indicative of eigenmoduli and eigenvalues of a plurality of transmission channels. 37. The receiver unit of claim 31, wherein the channel quality estimator further generates channel state information using unbiased minimum mean square error processing. 38. The receiver unit of claim 31, wherein the channel quality estimator further uses correlation matrix inverse processing to generate channel state information.

Images